Exploring the Impact of Internal Corporate Governance on the Relation Between Disclosure Quality and Earnings Management in the UK Listed Companies

Arcot and Bruno (2011) claim that disclosure and corporate governance are substitutive; hence the adherence to either one of these two components is basically effective in enhancing corporate performance. Opposing this view, Holm and Schøler (2010) point out that, corporate governance mechanisms are not perfectly substitutive for each other because the variation in corporate governance practices by firms is largely dependent on the unique needs and specific agenda of each firm.

Nonetheless, the Code does not suggest a specific number of board members. Paragraph B.1 the UK Corporate Governance Code (2010, p. 12) states that:. The board should be of sufficient size that the requirements of the business can be met and that changes to the board’s composition and that of its committees can be managed without undue disruption, and should not be so large as to be unwieldy.

In a similar vein, the US Blue Ribbon Committee (1999) recommended that audit committee meetings should be conducted not less than four times in a year.

We excluded companies ranked third (i.e. second runners-up) from our sample because a selection of control sample with multiple criteria might be problematic when the main sample is large; therefore, by using the winners and first runners-up, the selection of the control sample is more feasible and realistic.

Given that the awards covered multiple categories, a company could have received more than one award. The figure for our initial sample (see Table 2) refers to the number of non-unique companies receiving either the winner or the first runners-up award.

We conducted t-testing to check the mean differences of total assets in both winner and non-winner groups. Results show that there is a significant difference between the means of these two groups at p < 0.01. Nonetheless, it is argued that finding a perfect match is nearly impossible. Our finding is consistent with those of Peasnell et al. (2007) who used IRAWARD in the US as a proxy for investor relation activities. Specifically, in their match-paired sample, they acknowledge that there is a huge significant difference in firm size in the winner and non-winner groups (at p < 0.01). However, it is also worth noting that other criteria, such as industry and year, are used in determining the control sample. At least this helps to alleviate the weaknesses in the sample selection choice to a certain extent.

We traced the information on audit committee expertise in the directors’ profile section in the annual report. Following Hoitash et al. (2009, p. 848), we determined audit committee expertise if the audit committee member is holding any of the following (similar) qualification/position, namely: “certified public accountant; chief financial office;, principal financial officer; chief accounting officer; principal accounting officer; treasurer; auditor; vice president of finance”. Note that Hoitash et al. (2009) depend on the Securities and Exchange Commission (SEC) Final Rule when defining audit committee expertise in their research. We believe that our definition of audit committee expertise is in line with The UK Corporate Governance Code (2010) (Para 3.C.1) that “at least one member of the audit committee has recent and relevant financial experience” (p. 19).

We estimate non-discretionary accruals using Modified Jones Model suggested by Dechow et al. (1995), using the following formula: NDA _t  = α₁(1/LTA) + α₂(ΔREV _t  − ΔREC _t /LTA) + α₃(PPE _t /LTA), where NDA _t is the non-discretionary accrual in the year t divided by lagged total assets, ΔREV_t represent change in revenue in the year t (current year revenue minus last year revenue), ΔREC _t represent change in receivables in the year t (current year receivables minus previous year receivables) and PPE _t is the gross property, plant and equipment at the end of t. All components in the equations are divided by lagged total assets in order to reduce heteroskedasticity (Jones, 1991). We obtained the coefficient parameters (α₁, α₂ and α₃) by performing OLS Regression on at least 6 firms in each industry in consistent with Athanasakou et al. (2009), using this equation: TA/LTA = b₁(1/LTA) + b₂(ΔREV _t /LTA) + b₃(PPE _t /LTA) + e _t , where TA is total accrual and e is error term. We calculate the total accrual using cash flow approach following Jo and Kim (2007), where we subtract operating cash flow from earnings before extraordinary items and discontinued operations. According to Hribar and Collins (2002), the cash flow approach is superior to the balance sheet approach since the later suffers from serious measurement errors. After calculating the NDA, we then calculate the discretionary accrual (DA) by subtracting total accrual (TA/LTA) with DTA using the following equation: DA = TA − NDA.

Some of the forward-looking keywords used in Hussainey et al. (2003) include “accelerate, anticipate, await, envisage, estimate, eventual, expect, forecast, forthcoming, outlook and predict”.

The AFA is estimated as = (−1) |EPS _t  − MEPS_t|/PRICE _t , where EPS_t is earnings per share, MEPS _t is the median forecast of earnings per share and PRICE _t is the share price in period t, (share price at the beginning of the year).

This finding contradicts some earlier studies (e.g. Kent et al. 2010), but could be explained in several ways: (1) Some of the literature argues that high compliance with the UK Corporate Governance Code is merely due to “ticking the box” activities, while at the same time highlighting the importance of considering the various unique needs of each firm’s governance system (Arcot and Bruno, 2006; Siregar and Utama 2008; Arcot et al. 2010); (2) The effectiveness of an audit committee primarily depends on the effectiveness of the board of directors. Given that BOD characteristics (e.g. BODIND, BODSIZE, BODMEET) are insignificant in curbing earnings management (R ² increases only 0.0016 %), it is suggested that audit committees are not able to offer effective monitoring in the absence of the serious roles of the BOD in constraining earnings management; even though their composition, number of meetings, expertise and size are in compliance with the Smith Report (2003) and the UK Corporate Governance Code (2010). In other words, when monitoring by a board of directors is not helpful in reducing earnings management, it is not surprising to see that audit committees also fail to carry out effective monitoring functions, given that the latter is a subset to the former; and (3) Audit committees (where the majority of them are entirely comprised of external directors) mainly rely on the information prepared for them in order to provide necessary monitoring. They therefore have less information advantage as compared to internal directors (Adam and Ferreira 2007). It is very unlikely that internal directors will let external directors know that they have been engaged in earnings management (Armstrong et al., 2010), making it nearly impossible for external directors to detect such activity. For that reason, the compliance with audit committee characteristics as recommended by the Smith Report (2003) and the UK Corporate Governance Code (2010) might be, to some extent, useful in helping companies to structure their internal governance system; however, it is only marginally beneficial in constraining earnings management.